We identified six sets of cultural/contextual issues that require consideration in mhGAP implementation: (i) cultural differences in explanations of and attitudes toward mental disorders; (ii) the structure of the local health-care system; (iii) the level of supervision and support available post-training; (iv) the level of previous education, knowledge and skills of trainees; (v) the process of recruitment of trainees; and (vi) the larger socio-political context of the region.
Cultural differences in explanations of and attitudes toward mental disorders
Cultural knowledge, attitudes and practices exert strong effects on help-seeking, treatment referral, adherence, and response to interventions. Although trainees with local cultural knowledge are an essential resource in mental health service delivery, they present specific challenges to standardized mhGAP training: (1) trainees may share cultural beliefs, and assumptions with others in their local culture that lead to biases and stigma toward mental illness; (2) trainees may be unclear how to apply their cultural knowledge to the specific tasks of mhGAP; and (3) the local context may include particular types of clinical problems, presentations, and social predicaments not explicitly addressed in mhGAP.
Cultural attitudes toward mental disorders are important factors in service provision. Stigmatizing cultural beliefs, explanatory models and attitudes shared by both patients and mental health workers (e.g., the concern that mental disorders are contagious or involve supernatural causes that cannot be addressed by biomedicine) will shape service delivery. Lack of attention to cultural context on the part of providers and decision-makers can lead to mistrust of mental health information and services and reduce motivation to engage with mental healthcare or adhere to treatment [
43,
44]. In our experience, it is not uncommon to meet health practitioners who are convinced that mental disorder in a specific patient is due to curses, spirit possession, or to patients having behaved in ways that angered the ancestors. Some health providers who accepted the premise that psychotic symptoms had supernatural causes, believed this could confer immunity to physical illnesses. Such views among health care providers may affect interactions with patients and influence treatment choices.
Of course, the impact of cultural beliefs is not confined to mental conditions. Given that physical illnesses often provide signs that point to the presence of known biological agents, biomedical services may be more likely to be sought for physical conditions, but cultural meanings and implications of affliction remain a critical ingredient in the negotiation of care [
45]. In the absence of visible signs of disease, as is usually the case for mental disorder, there may be greater uncertainty about the nature of the affliction and its causes. A recent study in Uganda evaluating challenges to implementing the WHO guidelines for management of stress found that health care providers held that the psychological interventions developed in high-income countries were not adaptable to local settings in part because the guidelines did not address the cultural context [
46].
Cultural and contextual factors also influence the types of mental health problems that practitioners face. This may pose challenges to the applicability of mhGAP materials. For example, in the Ugandan studies, practitioners found the guidelines did not provide adequate tools for the treatment of comorbid disorders and did not address the types of psychosocial problems seen in the population including war-related and residual post-conflict difficulties of Ugandan population and the psychosocial difficulties particular to refugee populations, including, for example, those related to domestic violence and disputes over land [
47].
In our field work, trainees in other regions have expressed similar concerns. Examples included reported difficulty in managing cases of severe trauma, sexual violence, and child-marriages. Trainees working with refugee populations frequently expressed the need for specific training in inter-ethnic and conjugal conflict resolution. Trainees gave examples of instances they were called on to intervene in cases related to clashes among refugees, and inter-ethnic clashes between refugees and members of their host community; the latter often due to the perception that refugees were responsible for the use and depletion of common community resources such as water and fuel wood. Less common but equally challenging concerns were related to negative attitudes toward refugees held by some trainees themselves.
A large body of work demonstrates that cultural variations in illness experience and behavior influence the diagnosis and treatment of mental health problems [
48,
49]. Cultural knowledge, attitudes and practices exert strong effects on help-seeking, treatment referral, adherence, and response to interventions. People from many ethnocultural backgrounds do not seek mental health-care, either because they do not conceptualize their problems as appropriate for clinical attention or because they fear social stigma [
50]. Even when they do present for help, patients may receive incorrect diagnoses and inappropriate or inadequate help from care providers unfamiliar with their language, cultural background, or social situation [
51]. In the context of mhGAP, this may occur if generic methods are applied without adequate cultural adaptation and contextualization. While minimal adaptation is recommended to maintain fidelity to the evidence-based protocols and not compromise their efficacy, cultural and contextual adaption are necessary to account for local idioms of distress and patterns of help seeking behavior and to provide accurate tools for case-identification and referral to more specialized care [
52‐
54]. Crucially, cultural idioms of distress which provide everyday languages of suffering, usually cannot be simply mapped onto specific disorders but need to be explored in context case-by-case to determine the nature of the patient’s concerns [
55].
When mhGAP training is carried out in humanitarian settings, it is common for trainers and trainees to come from different cultures. In Chad, for example, mental health training has been provided by trainers from Canada, Democratic Republic of Congo, India, Belgium and the US. The trainees are French-speaking Chadian health-providers who treat Zaghawa-speaking Darfur refugees. In these settings, the historical, geopolitical, and social contexts play important roles in communication that may affect training. Local trainees may have much relevant knowledge about idioms and explanatory models but lack ways to integrate this into mhGAP delivery.
As discussed earlier, mhGAP training is based on a biomedical model and aims to be agnostic to health-care providers’ own ethnocultural backgrounds and assumptions [
40]. The biomedical model locates mental health problems within individuals and tends to downplay or discount social, moral, and spiritual explanations of distress [
56]. However, many local mhGAP-trained health-care providers may share the cultural models of their patients. Moreover, in primary care or community settings, mhGAP trainees often are called on to treat patients with ambiguous, vague or diffuse symptoms, which may be attributed to many causes [
57]. Trainees may have difficulty conceptualizing, diagnosing, and treating symptoms which suggest no clear etiology or are not grouped together with other symptoms in locally recognized syndromes.
The structure of the local health-care system
To be useful mhGAP training must fit the structure of the local health care system and the contexts of practice. Some studies have highlighted the importance of such structural factors in mhGAP implementation. For example, an evaluation of implementing mhGAP guidelines in Uganda [
47] cited the low ratio of provider to patients, lack of qualified staff, insufficient funding, lack of incentives for practitioners to modify their practice, lack of time for the addition of mental health evaluation to existing clinical work and a high level of ethnocultural diversity in patient populations as underlying barriers to adaptation and adoption.
In our work, we have observed structural challenges related to the logistics of service provision of two types: those related to tangible resources and those related to information resources. Both of these were frequently reported by trainees as obstacles to the integration of mhGAP training in their routine practice across settings including refugee camps, rural and urban health-care centers. Problems related to tangible resources included the availability of an adequate supply of psychiatric medications, and the availability of appropriate physical spaces for mental health consultations. Trainees working in very low resource settings were particularly concerned that the adoption of mhGAP could further burden the over-extended system without providing support or incentives for the increased workload.
Information resource challenges were associated with problems in communication and access. After training, most trainees received no feedback about the integration of mhGAP in the larger health care system or the specific impact of their work. Trainees frequently voiced concerns about a lack of information on how to access supervisory resources and referral pathways.
As previously mentioned, two essential components of mhGAP implementation are task shifting and the stepped care approach. These both depend on the structure of the health care system. The stepped-care approach depends on having appropriate referral pathways. Even if referral pathways are formally present, they may be infrequently used and function poorly in practice. Related to this, due to lack of a broader vision and strategy for mental health service systems in most LMIC, efforts to integrate mental health may be confined to Primary Health Centers (PHC). For example, in Ethiopia, health-care providers in local and regional hospitals were not given the requisite training to enable them to supervise PHC workers or handle patients who come through the referral system. An exclusive focus on one sector like primary care can contribute to fragmentation in the health-care system, with mental health-care sequestered in PHC and specialty mental health-care facilities found only in the larger cities, with little or no attention given to health-care providers in between, making it difficult to implement the stepped-care component of the mhGAP.
Other structural challenges that can impede efforts to implement mhGAP are related to the lack of mental health policy and planning. About half of the countries in Africa do not have mental health policy, and among those that do, nearly 40% have not updated or modified their policies since 1990 [
37]. For example, in Kenya, the lack of comprehensive mental health policy has impacted on the ability to coordinate, evaluate and standardize service provision within the mental health system [
58]. In Chad, our experience suggests that, while the government recognizes the need for mental health training for health service providers throughout the country, this responsibility is largely directed by international humanitarian agencies who have mainly focused on the refugee population. Another common situation is that efforts in scaling-up mental health services emerge as a reaction to dramatic health events or crises like epidemics or natural catastrophes [
59‐
61]. This is important because mhGAP training programs are best implemented within a wider national mental health plan and policy and a cohesive mental health system.
The level of supervision and support available after training
In our field experience, mhGAP training without follow-up supervision generally has been insufficient to ensure the integration of mental health knowledge into practice. Many difficulties in the application of mhGAP are related to errors in differential diagnosis leading to inappropriate care. In most cases, pre-post tests indicate increases in mental health knowledge with mhGAP training and, in some instances, post-training evaluations indicated that most trainees had high levels of confidence in their ability to administer mental health-care services. However, there is some indication from the same study that high levels of confidence may not correlate with post-training knowledge test scores [
62]. In training sessions in which we participated, many practitioners focused on detecting psychotic symptoms and gave a diagnosis of psychosis without sufficient evidence. In other instances, trainees had difficulty in assessing symptoms in children and the elderly. Examples include trainees giving a diagnosis of psychosis to a child with developmental delays and to an elderly man with symptoms of dementia. Non-psychotic behaviors such as aggression, agitation, motor restlessness, and wandering may be routinely attributed to psychosis. Lack of experience in clinical interviewing and evaluation and insufficient history-taking may contribute to potential misdiagnosis.
In addition to the challenges of misdiagnosis, with over- or under-recognition of specific conditions, lack of supervision can also affect the kind and quality of interventions practitioners provide. In our experience in Ethiopia (mhGAP training 2015–2016), in the absence of adequate supervision, primary health-care physicians may be reluctant to prescribe medications in adequate doses because of concern about side effects. Diagnostic assessment and treatment approaches are mutually reinforcing. In the absence of supervision, health workers may attempt to fit symptoms they observe into the few broad categories of mental disorder they have learned about in the mhGAP program. This may result in over-diagnosis of some illnesses and potential over-prescription, thereby exposing patients at least to unnecessary expense and medication side-effects [
63]. Under-diagnosis and under-treatment can also be observed, when the skills that have been taught in the classroom do not sufficiently generalize to routine clinical practice. More research and evidence are needed to understand the factors that support the effective translation of mhGAP training into practice at the primary care level.
The level of previous education, knowledge and skills of trainees
As discussed above, mhGAP aims to narrow the service gap by task shifting (i.e. having primary-care providers offer mental health-care). Mental health training of healthcare providers varies considerably across countries and regions [
64]. There is some evidence that in LMIC, mhGAP-trained health providers without prior mental health training or those with low levels of formal education can successfully assess patient needs and provide support when using locally validated measures [
65]. However, trainee variation in education and skill levels can pose challenges for mhGAP training as the training program assumes a level of general knowledge that may not be consistently present among workers [
66]. Insufficient basic health education may leave trainees with biases and misinformation about mental disorder. Lack of adequate knowledge and training is a common deterrent for mental health service providers in LMIC as they do not feel competent to care for people diagnosed with mental disorders [
66]. Although mhGAP aims to reduce common assumptions that mental disorders are contagious or caused by witchcraft, it may not mitigate stigma. Some workers may harbor negative attitudes toward mental disorders that remain unchanged by their exposure to mhGAP training. We have observed that some trainees maintain the lay belief that mental disorder is contagious because they misidentify infectious diseases. For example, when asked about the origin of this notion of contagion, a nurse trainee recounted the case-history of a patient with meningitis.
Delivery of mhGAP training includes pre- and post-tests of mental health knowledge. These tests are used in almost all mhGAP training courses including refresher and train-the-trainer initiatives. Ideally these tests could provide important feedback on the success of training. However, to our knowledge, the test questions have not gone through an evidence-based process of item construction and the pre- and post-tests are not standardized measurement instruments. Their cross-cultural validity has not been established. We have observed trainees with different educational professional backgrounds and skill-levels have difficulty with some of the assessment questions. While there is an overall trend toward improvement at post-test [
62], some trainees actually scored lower after the completion of the training compared to the start of the training. This may reflect trainees’ new-found appreciation for the subtleties of clinical work.
Recruitment of trainees
Much preparation is needed prior to implementation of mhGAP training. A key logistical element that is not always considered systematically involves choosing and inviting trainees to enroll in the mhGAP training program. The process of selecting trainees is crucial to the ultimate success of mhGAP training. In our experience, relegating the task of trainee selection to beneficiary organizations or the state, may not lead to the best choice of candidates for mental health training, or the best outcomes in terms of accessible mental health-care for the underserved populations.
At the level of the clinic, we have observed that mhGAP trainees may be selected based on status and standing in their respective organizations. In some instances, this may even result in health administrators being recruited to the exclusion of clinical staff. While health administrators can certainly benefit from mhGAP training and can play a role in future training and supervision of clinical staff, their increased competence may not translate into better patient care because administrators are usually released from clinical duties. Status-based selection is also often tied to gender and cultural defined norms. Thus, female nurses, midwives, and birth attendants may be overlooked in the selection process in localities where women have lower standing in the work place or the community.
At the grassroots level, CHWs are recruited and trained as peer-support workers, sometimes without clear criteria for inclusion. This group represents an important human resource in LMIC, but recruitment strategies may ignore existing help-seeking pathways, traditional healers, and other sources of care that can hinder or promote mental health interventions. The most effective CHWs may be those who exert the most influence in their communities, who are respected opinion leaders, and who are centrally positioned in the community and able to provide information and assistance to as large a group of people as possible.
Historical and current political context
The political context, both historical and current, of the community, region, and nation where mhGAP training is undertaken influences local health systems and the population’s willingness to adopt new health practices. In West Africa, for example, histories of political conflict and violence eroded trust in government institutions, including health care, with devastating effects on the response to the Ebola epidemic [
67,
68]. Public health efforts directed at the epidemic in West Africa were met with reticence and resistance from the population in most affected regions but especially in Guinea a “crisis in confidence” [
69] resulted in great resistance to public health efforts compared to that of its neighbors (Sierra Leone and Liberia) [
70,
71]. The public’s fear of government systems also affected efforts at mental health system implementation and strengthening, including mhGAP training and service delivery, making it difficult to respond to mental health needs even when services were available. The propagation of misinformation, and incidents of violence directed to heath workers reflected entrenched fears of public officials, including health care workers [
72]. Perhaps as a result, the service rooms of the clinics of Fraternité Medical Guinea (FMG), Guinea’s only health clinics offering mental health services by mhGAP trained physicians, remained largely empty. In what incident we observed, after mhGAP training, healthcare workers’ attempt to refer an Ebola survivor with severe mental disorder [
73] for admission to the Donka hospital in Conakry, which houses the only public psychiatric service unit in the country was fraught with difficulties borne of the public’s deep-rooted distrust of government and authorities. Challenges
2 included the family’s reluctance to consent to the patient’s medical transport to the hospital, an attack by a mob that tried to torch the inter-city ambulance transporting the patient, and the patient’s multiple attempts to flee the hospital he distrusted.
Although these were extraordinary circumstances, issues of trust and engagement with government or NGO programs are common in LMIC with histories of colonization, political instability and violence. There has been little discussion of this type of historically rooted challenge in the literature on public health efforts and mhGAP program implementation. However, these contextual factors play an important and largely implicit role in the adoption of programs such as mhGAP, both in the response to training, and in its integration in service provision. Strategies are needed to explore and address historically rooted political issues that may impede program implementation.